REAL EXCHANGE RATE EFFECTS ON THE BILATERAL MANUFACTURING TRADE BALANCE BETWEEN THE US AND TÜRKİYE: A BOUNDS TESTING APPROACH WITH STRUCTURAL BREAKS

Abstract

The exchange rate–trade balance relationship occupies a central place in open-economy macroeconomics, yet empirical evidence on whether currency depreciation generates a J-curve adjustment remains sensitive to the level of aggregation at which trade data are measured. Sector-level analysis is particularly informative in this regard, as it avoids the aggregation bias that arises when opposing dynamics across different product categories cancel each other out in country-level trade balance series. This paper investigates dynamics in the bilateral manufacturing trade balance between the United States and Türkiye using monthly sector-level data over the period 2013M01–2025M12. The trade balance is defined as the ratio of Türkiye's manufacturing exports and Türkiye’s manufacturing imports to the US. Structural breaks are identified endogenously through the Lee–Strazicich (2003) unit root test, and a step dummy variable control for the permanent level shift associated with the COVID-19 shock from March 2020 onward. The ARDL bounds test confirms a stable long-run cointegrating relationship, and the error correction estimates reveal a net negative short-run exchange rate effect that reverses to a positive and significant long-run effect, consistent with the J-curve hypothesis. The findings suggest that real depreciation of the Turkish lira improves the bilateral manufacturing trade balance in the long run, though a temporary deterioration should be anticipated before the competitiveness gains materialize.